Beruflich Dokumente
Kultur Dokumente
Version 1.0
March, 2014
Page 1
Executive Summary:
This guidance document represents the culmination of a multi-year joint project between the
Colorado Department of Public Health and Environments Water Quality Control Division (the
Department) and the Colorado State Universitys (CSU) Department of Civil and Environmental
Engineering. The document is meant to provide technical assistance to small public water
systems with regard to disinfection, disinfection efficacy, and maximizing their ability to achieve
and comply with disinfection as required by regulations. This guidance document presents a
number of pre-engineered small-scale tanks (less than 5,000 gallons and operating at up to 50
gallons per minute (GPM)) and tank/pipe configurations that can be used for chemical
disinfection as part of a drinking water treatment system. All of the pre-engineered systems and
system modifications have undergone extensive research and testing at the Engineering Research
Center at the Colorado State University. Specifically, this research study investigated
disinfection through pipe segments, pressurized storage tanks, non-pressurized plastic storage
tanks, and non-pressurized rectangular concrete tanks. Additionally, this research investigated
inlet manifolds and random packing material as potential modifications to new or existing
disinfection tanks. In essence, this guidance manual outlines the expected baffling factor (BF) at
a range operational flow rates for these pre-engineered systems and modifications. The work
presented in this document was initiated by the Department with the intention that small drinking
water systems will be able to utilize the results that are presented in this guidance document by
installing one of the pre-engineered small-scale tanks and/or implementing the recommended
modifications to their existing infrastructure in order to comply with treatment requirements. The
table below presents a summary of the pre-engineered systems and modifications presented in
this document.
Summary of pre-engineered systems and modifications presented in this Guidance
Document
Section
Number
2
3
4
5
6
7
System/Modification Type
Pipelines and Pipe Segments
Pressurized Retention Tanks
Open Surface Concrete Tanks
Non-pressurized Plastic Tanks
Inlet Manifolds*
Packing Material*
Range of
BF
0.6-1.0
0.1-0.5
0.1-0.5
0.1-0.2
0.1-0.5
0.1-0.6
Flow Rates
(GPM)
5
5-30
20
<50
<50
<50
* Sections 6 and 7 are modifications that can be used to increase the baffling factors of the disinfection
tanks discussed in Section 5.
Page 2
Acknowledgements
The Colorado Department of Public Health and Environment gratefully acknowledges the
individual contributions of the following:
Page 3
Contents
1 INTRODUCTION ........................................................................................................................................... 6
1.1PURPOSE OF THE THIS GUIDANCE MANUAL ....................................................................................... 6
1.2 WHAT ARE BAFFLING FACTORS HOW ARE THEY MEASURED OR ASSIGNED? ......................... 7
1.3 CT AND LOG INACTIVATION ................................................................................................................. 9
1.4 RESEARCH PERFORMED BY COLORADO STATE UNIVERSITY ...................................................... 10
1.5 LOCATION OF RESULTS OF MODELING AND BAFFLING FACTOR TRACER TESTING .............. 11
2 BAFFLING FACTORS FOR PIPELINES AND PIPE SEGMENTS ................................................................... 13
2.1 BAFFLING FACTORS AWARDED FOR PIPELINES AND PIPE SEGMENTS ...................................... 14
2.2 CONSTRAINTS FOR PIPELINES AND PIPE SEGMENT SYSTEMS ..................................................... 14
2.3 SYSTEMS DESIGNED OUTSIDE THE CONSTRAINTS......................................................................... 14
2.4 EXAMPLE: PIPE LOOP CONTACTOR CALCULATION ....................................................................... 15
3 BAFFLING FACTORS FOR PRESSURIZED RETENTION TANK SYSTEMS ............................................... 17
3.1 BAFFLING FACTORS AWARDED FOR PRESSURE TANKS PLUMBED IN SERIES ......................... 18
3.2 CONSTRAINTS FOR PRESSURE TANK SYSTEMS ............................................................................... 18
3.3 SYSTEMS DESIGNED OUTSIDE THE CONSTRAINTS......................................................................... 18
3.4 EXAMPLE: PRESSURE TANK CONTACTOR CALCULATION ........................................................... 20
4 BAFFLING FACTORS FOR OPEN SURFACE CONCRETE TANKS ............................................................. 21
4.1 BAFFLING FACTORS AWARDED FOR OPEN SURFACE CONCRETE TANKS ................................ 23
4.2 CONSTRAINTS OF OPEN SURFACE CONCRETE TANK SYSTEMS .................................................. 24
4.3 SYSTEMS DESIGNED OUTSIDE OF THE CONSTRAINTS ................................................................... 24
4.4 EXAMPLE: OPEN TANK CONTACTOR CALCULATION ...................................................................... 26
5 BAFFLING FACTORS FOR NON-PRESSURIZED NON-CONCRETE (PLASTIC) TANKS ........................ 28
5.1 BAFFLING FACTORS AWARDED FOR NON-PRESSURIZED TANK SYSTEMS .................................. 28
5.2 CONSTRAINTS FOR NON-PRESSURIZED PLASTIC TANK SYSTEMS .............................................. 29
5.3 SYSTEMS DESIGNED OUTSIDE THE CONSTRAINTS......................................................................... 29
5.4 EXAMPLE: OPEN TANK CALCULATION.............................................................................................. 30
6 BAFFLING FACTORS FOR INLET MANIFOLDS .......................................................................................... 31
6.1 BAFFLING FACTORS AWARDED FOR MANIFOLDS ........................................................................... 32
6.2 CONSTRAINTS FOR MANIFOLD SYSTEMS .......................................................................................... 33
6.3 SYSTEMS DESIGNED OUTSIDE THE CONSTRAINTS......................................................................... 34
6.4 EXAMPLE: INLET MANIFOLD CALCULATION ................................................................................... 34
7 BAFFLING FACTORS WITH TANK PACKING MATERIAL ........................................................................ 36
7.1 BAFFLING FACTORS AWARDED FOR PACKING MATERIAL ........................................................... 36
7.2 CONSTRAINTS FOR PACKING MATERIAL.......................................................................................... 38
7.3 SYSTEMS DESIGNED OUTSIDE THE CONSTRAINTS......................................................................... 38
7.4 EXAMPLE: TANK PACKING MATERIAL ............................................................................................. 39
ORIGINAL BF ..................................................................................................................................................... 40
APPENDICES ................................................................................................................................................... 41
APPENDIX A LOG INACTIVATION CALCULATIONS ................................................................................ 41
APPENDIX B PHASE 2 REPORT ....................................................................................................................... 42
APPENDIX C EVALUATION OF FLOW AND SCALAR TRANSPORT ............................................................ 43
Page 4
Page 5
Page 6
readily available from major water industry parts distributors, to ensure compliance with surface
water treatment and ground water rules stipulated in CPDWR. To this end, the multi-year joint
project between the Colorado Department of Public Health and Environments Water Quality
Control Division (the Department) and the Colorado State Universitys (CSU) Department of
Civil and Environmental Engineering focused on developing several cost effective preengineered tank configurations applicable for use by small drinking water systems. This
Guidance Manual is the culmination of this effort between the Department and CSU and it
provides guidance on how to calculate detention times, appropriate baffling factors, and how to
assign disinfection credit for the pre-engineered systems that were studied at CSU. The following
systems were studied:
pipe segments,
pressurized tank systems,
non-pressurized storage tank systems,
rectangular tanks with serpentine baffles
The Guidance Manual also covers cost effective recommendations that are designed to increase
the baffling factor of small contact tanks. In this guidance document, a contact system (or more
precisely a disinfection segment) is any tank or pipe system used to achieve contact time
between a disinfectant and raw water (incoming supply). Additionally, this Guidance Manual
provides guidance on several types of system modifications that can be used to increase the
baffling factor of both new and existing small systems.
All of the pre-engineered systems and system modifications have undergone extensive physical
testing as well as computational flow modeling. As such, this Guidance Manual outlines the
expected baffling factor and operational flow rates for these pre-engineered systems and
modifications. At the end of each subsequent section (Sections 2 through 7), an example on how
to calculate the contact time to determine the log inactivation of the studied disinfection system
is provided.
1.2 What are Baffling Factors how are they measured or assigned?
The baffling factor BF of a contact tank is used to adjust the theoretical detention time to a
more realistic value of the CT of the system which has been defined earlier. A reliable and
accurate method to determine the BF of a disinfection system is through the use of a tracer study.
During a tracer study, a tracer chemical is injected into the influent. This injection point should
be as close as possible to the disinfectant injection port. The water containing the tracer chemical
travels through the contact volume, then the concentration of this tracer is monitored at the outlet
over time. A resident time distribution (RTD) curve is then generated by plotting the
concentration of tracer at the chlorine contact system outlet as a function of time. Figure 1 shows
an example of a RTD curve of a step dose tracer input for a hypothetical contact system. This
RTD curve would be associated with a moderately efficient disinfection segment and would have
Page 7
a BF (=T/TDT) of 0.5, indicating that the flow short circuits through the disinfection segment. In
contrast, the plug flow line shown in Figure 1 depicts the case when all of the tracer material sent
through the disinfection segment reaches the outlet at the theoretical detention time (TDT) of the
segment. For a detailed discussion on tracer studies please refer to Section 2.6 in the Phase 2
Final Report provided in Appendix B and the tracer studies protocol prepared by the
Department that is provided in Appendix I.
Page 8
Therefore, this current Guidance Manual expands on and clarifies the baffling factors listed in
Table 1 in order to ensure that appropriate baffling factors are assigned.
Table 1: Baffling Factors
Baffling
Condition
Unbaffled
(mixed flow)
Baffling
Baffling Description
factor
None, agitated basin, very low length to width
0.1
ratio, high inlet and outlet velocities
Poor
0.3
Average
0.5
Superior
0.7
Perfect
(plug flow)
1.0
volume, V of the disinfection segment (e.g. a contact tank), peak hourly flow rate Q, and the
baffling factor BF. The volume of the disinfection segment should be the lowest volume that
occurs when the treatment system is operational, while the peak hourly flow rate should be the
highest hourly flow rate that occurs during operation. For equations to calculate the volume of
various shapes please refer to Table 2. The theoretical detention time TDT is the contact time
of the system if it had perfect plug flow and is calculated using Equation 1. Plug flow occurs
when the water flows through a contactor such that all water remains within the vessel exactly
the TDT and no short-circuiting occurs. BF is used to adjust the TDT so that it reflects the actual
flow conditions within the tank. A tank with a BF of 0.1 would have a high amount of shortcircuiting, dead zones, and recirculation, while a system with a BF of 1.0 indicates ideal plug
flow conditions. Equation 2 shows how the contact time T is calculated from the TDT and the
BF.
(1)
(2)
For more information on how CT is calculated, including a step-by-step example, please see
Appendix A. This appendix contains the Departments brochure, which provides additional
details on calculating log inactivation.
Table 2: Volume Equations for Shapes
Shape
Volume equation
Pipe
Rectangular
Basin
Length Width
Minimum Water Depth
Cylindrical
Basin
Length
Cross-sectional
Area (r2)
please see Appendix I. Both purpose built laboratory systems and existing water treatment plants
were tested using physical tracer studies.
Specifically the types of systems tested included:
Pipe Segments
Pressurized Tank Systems
Non-Pressurized (NP) Plastic Tank Systems
Concrete Tank Systems
In addition to the systems tested, several types of tank modifications were also tested. These tank
modifications included:
Inlet Manifolds
Random Packing Material
In what follows in this Guidance Document, each of the above systems and modifications are
discussed in Sections 2 through 7. Each of these sections provides an overview of the studied
system followed by the appropriate baffling factors, constraints and an example calculation of
the contact time. Please note that the material presented in these sections is the culmination of a
five year research effort that involved extensive laboratory and field experiments as well as CFD
simulations. Therefore, each section will refer the reader to the appropriate appendix to justify
the assumptions and data that back up the assigned baffling factors.
1.5 Location of Results of Modeling and Baffling Factor Tracer Testing
For more detailed information of all of the studies conducted, please refer to:
Page 11
APPENDIX D: Pipe Loop CFD Models. This appendix contains the results of a CFD
modeling study that aims to highlight the differences in baffling factor for laminar and
turbulent flow in pipe loops as well as the minimum pipe length to diameter ratios
required in pipe loop contactors to achieve near plug flow conditions.
APPENDIX E: Open Surface Concrete Tanks. This appendix contains the results of
physical tracer studies that investigated how to improve the baffling factor of concrete
tanks using baffles and random packing material.
APPENDIX F: Flow Dynamics and Scalar Transport in Drinking Water Contact Tanks.
This appendix contains a Master of Science thesis presents the results of CFD modeling
and tracer studies conducted on serpentine baffle tanks, non-pressurized plastic storage
tanks, inlet modifications, and use of random packing material, as well as a case study of
an operational disinfection segment.
APPENDIX G: Towards Improved Understanding and Optimization of the Internal
Hydraulics of Chlorine Contact Tanks. This appendix contains a Master of Science thesis
presents results of CFD modeling and tracer studies conducted on serpentine baffle tanks
and inlet manifolds.
APPENDIX H: Improving Drinking Contact Tank Hydraulics using Random Packing
Material. This appendix refers to a peer-reviewed journal article that is forthcoming in
February 2014 in the Journal of American Water Works Association (AWWA). It
presents results from tracer studies of vertical cylindrical tanks that are packed with
random packing material to improve the baffling factor.
APPENDIX I: Tracer Study standard operating procedure (SOP). This appendix prepared
by the Department outlines how to conduct a tracer study and shows the results of 2
tracer studies conducted by CSU.
These documents are included as part of this document or attached via hyperlinks in the
appendices with the exception of Appendix H. Appendix H is available for purchase from the
Journal of the American Water Works Association.
Page 12
Minor Length
Major Length
To develop turbulent flow and be awarded a BF of 1.0, flow rate must be above the
minimums listed in Table 3; there is no maximum operating flow rate except as limited
by the required contact time.
The pipe must maintain a constant diameter throughout (i.e. there may be no expansions
or contractions).
System must have a total L/D ratio 160.
Page 14
For piped systems with an L/D <160 but with an L/D 40 overall, a BF of 0.7 will be assigned
provided the flow conditions are turbulent as stipulated in Appendix D.1 (i.e., Reynolds Number
of the flow must be greater than 4000). Any system that does not have an L/D 160 must
conduct a tracer study or otherwise justify the appropriate BF. For pipe loops with multiple
disinfection segments, a tracer study may be necessary or more conservative baffling factor
assigned if the individual runs of a pipe loop do not have an L/D ratio 40.
If a larger pipe diameter is to be used than those listed in Table 3, please refer to Appendix D.1
on how to determine minimum flow rates.
If the system does not have a constant pipe diameter, a tracer study will be needed to determine
the actual BF of the system. A tracer study is required because the sudden expansion/contraction
caused by a change in pipe diameter induces non-plug flow phenomena within these transitional
sections.
2.4 Example: Pipe Loop Contactor Calculation
4 inches
0.333 feet
2 inches
0.167 feet
15 feet
6
25 GPM
Page 15
From Table 3, the volume of the pipe loop contactor shown in Figure 3 is:
The system in this example would have a BF of 1.0 with a T of 2.3 minutes. This pipe loop
would have a total L/D of 273. Since this is larger than the required minimum of 160 this system
would receive the full BF credit of 1.0.
Page 16
Page 17
Inlet and outlet must be located on either the top (top side) or bottom (bottom side) of the
tank.All tanks must remain full to achieve the desired BF.
Both inlet and outlet cannot be located on same end of tank (e.g. top/top or
bottom/bottom), instead they must be located on opposite ends (e.g. top/bottom or
bottom/top)
Single inlet/outlet pressure tanks (e.g. bladder pressure tanks) will not receive any credit
Orientation (vertical as tested or horizontal) of tank does not alter BF.
Tanks must be plumbed in series if more than one will be used (i.e. the effluent of one
tank must flow into the next tank).
Total combined volume of all tanks in a series must not exceed 600 gallons. Any system
with a combined volume over 600 gallons was not verified through testing or modelling
and must conduct a tracer study or otherwise justify the appropriate BF.
Page 18
expect very high-pressure drops across the contact system. These pressure losses occur through
the entire system (e.g. piping, valves, tanks etc.) and numerous system modifications would be
needed to mitigate the losses.
If tanks are configured (plumbed) in a manner where the inlet and outlets are plumbed into the
sides of the tank and are not located at opposite ends of the tank, a maximum of 0.1 BF credit
will be issued. Tracer studies may be used to determine an alternate BF of the system. This
document will not automatically issue any higher credit to these systems due to the increased
short-circuiting that would be caused by such inlet/outlet configurations. This increased shortcircuiting would significantly reduce the BFs that were found in the CSU study.
If a system uses tanks where both the inlet and outlet are located on the same end of the tank
(e.g. top/top or bottom/bottom), increased short-circuiting would be caused by these inlet/outlet
locations which would significantly reduce the BFs found in the CSU study. The BF for such
systems cannot be definitively identified without the use of a tracer study.
If a system uses a bladder pressure tank (or tanks that use a shared inlet/outlet), no credit will be
issued. These tanks do not allow constant flow through the tank and are instead intended for
maintaining constant pressure within the system.
Page 19
Number of Tanks
Tank Volume
Flow Rate
3
80 gallons
15 GPM
From Table 4 the BF of a 3-tank system will be 0.3. The contact time (T) is then calculated by:
The system in this example would have a BF of 0.3 with a T of 4.8 minutes.
Page 20
Page 21
Plan View
Figure 7: Un-modified Tank (
Plan View
Figure 8: Tank with Inlet Box (
)(
GPM 1.5 diameter;
Page 22
Section View
)
Section View
GPM - 2 diameter inlet;
GPM 1 diameter)
Plan View
Figure 9: Baffled Tank (
Plan View
Figure 10: Baffled Tank - Inlet Box (
GPM 1.5 diameter;
Section View
)
Section View
)(
GPM - 2 diameter inlet;
GPM 1 diameter)
Plan View
Section View
Figure 11: Baffled Tank - Inlet Box w/Turn Boxes (
)
(
GPM - 2 diameter inlet;
GPM 1.5 diameter;
GPM 1 diameter)
4.1 Baffling Factors Awarded for Open Surface Concrete Tanks
Table 5: BF for Open Surface Concrete Tanks
Tank Setup
BF
Un-modified Tank
Tank with Inlet Box
Baffled Tank (per Figure 9)
Baffled Tank with Inlet Box
Baffled Tank with Inlet Box and Turn Boxes
0.1
0.2
0.3
0.4
0.5
Page 23
The inlet box length must be at least one third of the tank length (
Page 24
prints will behave differently than the tested systems which support this Guidance Manual (see
Open Surface Concrete Tanks in Appendix E).
Baffling factors for these systems must be obtained through the use of outside literature or
through independent physical tracer studies. The study titled Improving Clearwell Design for CT
Compliance (Crozes et. al. (1999), AWWA Research Foundation) has been widely cited to
justify baffling factors at larger scale water treatment facilities.
Tracer studies can be performed using procedures outlined in Appendix C, specifically in Section
2.4 in Evaluation of Flow and Scalar Transport Characteristics of Small Public Drinking Water
Disinfection Systems Using Computational Fluid Dynamics. An example standard operating
procedure (SOP) for performing a tracer analysis is provided in Appendix I.
Alternate Baffle Configuration: Placing baffles parallel to the longest axis of a given tank
maximizes the length to width ratio of the resulting channels and reduces the amount of flow
separation by limiting the number of turns. Setting the baffle opening (
) equal to the channel
width ( ) also reduces flow separation by avoiding contraction or expansion of the flow area.
Maximizing the length to width ratio and reducing the amount of flow separation encourages
fully developed flow within the contact tank, increasing its hydraulic disinfection efficiency.
Numerical justification for these concepts can be seen in Appendix G, specifically Section 3.6 in
Towards Improved Understanding and Optimization of the Internal Hydraulics of Chlorine
Contact Tanks. A minimum of 2 baffles within the contact tank is required in order to mitigate
the jet caused by the sharp inlet, otherwise fully developed flow does not occur and the effects of
the baffle are lost (see Appendix E). For alternate baffle configurations, a tracer study must be
performed to determine the BF for the tank system.
Non-Rectangular Geometry: Tanks with non-rectangular foot prints will behave differently than
the systems which supported this Guidance Manual and should thus be considered separately
(see Appendix E). It is unlikely that the application of internal baffling and random packing
material to tanks with non-rectangular geometry will yield similar results to the rectangular
contact tanks described in this section. If a tank is circular and conforms to respective
constraints, modifications recommended in Sections 6 and 7 of this document may be applied.
Otherwise, baffling factors of non-rectangular systems must be determined through the
application of tracer studies.
Operating Flow Rates Below 20 GPM: Small systems that use a 2-inch diameter pipe at
flowrates below 20 GPM approach a regime change at which inlet velocities are too low to be
constructively dissipated by the proposed application of random packing material. At 20 GPM,
the flow approaches a transition between turbulent and laminar flow regimes. Introducing
random packing material benefits the system by dissipating the high velocity jet introduced by
Page 25
the use of a sharp inlet, which decreases short-circuiting and disperses the incoming flow. As the
flow rate drops so does the amount of energy contained in the incoming jet. When the inlet flow
is below 20 GPM, the amount of energy in the incoming jet is low. The enhanced dispersion in
such cases is negated. Therefore, the minimum operating flow rates are as follows:
For 2-inch pipe: 20 GPM
For 1-inch pipe: 12 GPM
For 1-inch pipe: 5 GPM
The above flow rates preserve the inlet velocity allowing dispersion of the jet. See Open
Surface Concrete Tanks in Appendix E for additional justification. It is recommended that
internal baffling be used in these cases if an increase in baffling factor is desired (see Appendix
G, specifically Section 3.5 in Towards Improved Understanding and Optimization of the
Internal Hydraulics of Chlorine Contact Tanks or Appendix F, specifically Section 3.3 in Flow
Dynamics and Scalar Transport in Drinking Water Contact Tanks).
Alternate Inlet/Outlet Configurations: A system that is plumbed such that the influent enters at
the top of the tank and the effluent leaves at the bottom of the tank will receive similar benefits
from baffling as a system where the inlet enters at the bottom and the outlet exits at the top of the
tank. However, the application of packing material in this tank setup would require that the
material to be suspended and the resulting flow dynamics would not behave similarly. Tracer
studies must be used to determine baffling factors in systems attempting to use random packing
material with plumbing that differs from the prescribed conditions.
Inlet Box and Turn Boxes: Geometric constraints placed on the design of inlet and turn boxes are
based off of studies outlined in Appendix E and should not be modified.
Plan View
Section View
Baffled Tank
1,500 gallons
25 GPM
From Table 5 the BF of a baffled tank as shown in the schematic in Figure 12 will be 0.3. The
contact time (T) is then calculated by:
Page 27
Page 28
Page 29
Doorway Tank
500 gallons
25 GPM
From Table 6 the BF of a doorway tank shown in Figure 15 will be 0.2. The contact time (T) is
then calculated by:
Page 30
Page 31
Page 32
Plan View
Section View
Page 33
Section View
Plan View
Figure 19: Inlet Manifold System Used in Example
Original BF
Page 34
0.1
500 gallons
16
0.5
6 feet
30 GPM
16.7 minutes
1.7 minutes
4 feet
The system in this example (Figure 19) would have a sixteen-inlet manifold that is 7.2 inches
above the tank floor. This inlet manifold would give the tank a final BF of 0.5 and a new contact
time of 8.35 minutes, which will increase the contact time by 500%.
Page 35
void ratio of this material will not impact the performance. The void ratio is the volume of empty
space divided by the total volume of the material.
Table 8: Revised BF and Percent of Available Tank Volume with Random Packing
Material added to an open tank with a BF of 0.1 (see Section 4 and 5 for Open Tank BFs)
Percentage of Tank Filled with
New BF
Percent of Available
Packing Material
of System
Tank Volume*
0%
0.1
100%
25%
0.2
95%
50%
0.3
90%
75%
0.45
85%
100%
0.6
80%
*Note: This assumes the packing material has a void fraction of 0.8. Packing material tested at
slightly higher void ratios (0.90-0.95) provide higher baffling factors (see Appendix H.1,
specifically the journal article referenced therein). Please refer to Appendix H.2 for equation to
calculate Percent of Available Tank Volume that can be used for materials with void fractions
other than 0.8.
Once the amount of packing material to be used has been decided the useable tank volume
and the theoretical detention time (TDT) needs to be recalculated. The volume of the
tank will decrease due to the fact that the packing material takes up a certain percentage of the
volume. In order to calculate the new
and TDT, the following will be needed:
The new
Page 37
Plan View
Section View
Only random packing material that is 4 inches in diameter or smaller can be used.
Tank volume cannot exceed 500 gallons.
Tank flow rate cannot exceed 50 GPM
Tanks must be plumbed such that the influent enters at the bottom and the effluent leaves
at the top of the tank as shown in Figure.
Page 38
If a system uses packing material with a diameter greater than 4 inches, a tracer study must be
used to develop the system BF. See Appendix F, specifically Section 5.4.1 of Flow Dynamics
and Scalar Transport in Drinking Water Contact Tanks or Appendix H.1 for justification. It
must be noted that packing material smaller void ratios (less than 0.8) imply significant reduction
in useable tank volume and hence might not be practically useful for this purpose. Higher void
ratios up to 0.95 have been tested and can be used. However, no additional baffling factor credit
will be provided to those shown in Table 8.
If a system reverses the flow direction from that shown in this section of the guidance document
(see Figure 21), the system can only receive the improved BF if the tank is completely full of
packing material. When the contact tank is completely full of packing material, the system will
be credited a BF of 0.6 and the available tank volume will still need to be reduced as shown in
the example below. In this case the only difference would be the variable Vtank should be the
lowest system volume experienced during normal operation. For justification please see
Appendix H.1.
7.4 Example: Tank Packing Material
Page 39
Original BF
Tank volume w/o packing material
0.1
75%
0.45
85%
10 GPM
0.5 minutes
50 gallons
The system in this example (shown in Figure 22) will have a final BF of 0.45 and a new TDT of
4.25 minutes, which will increase the contact time from 0.5 minutes to 1.9 minutes. This is a
380% increase in contact time from the original tank.
Page 40
APPENDICES
APPENDIX A Log Inactivation Calculations
Page 41
Link
Page 42
Evaluation of Flow and Scalar Transport Characteristics of Small Public Drinking Water
Disinfection Systems Using Computational Fluid Dynamics
MS Thesis by Jordan M. Wilson (2011), Department of Civil and Environmental Engineering,
Colorado State University
Advisor Professor Karan Venayagamoorthy
This thesis focuses on the evaluation of flow and scalar transport characteristics of small
disinfection systems, primarily through computational fluid dynamics (CFD) as well as physical
conservative tracer studies. Original research was performed on a pipe loop, series of pressurized
tanks, and two separate open surface tank contact systems and a case study was performed on a
baffled tank system. The flow dynamics for each of these respective disinfection systems were
evaluated using CFD. Research presented in this study using CFD models and physical tracer
studies shows that evaluation methods based upon TDT tend to overestimate, severely in some
instances, the actual hydraulic efficiency as obtained from the systems' flow and scalar transport
dynamics and subsequent RTD curves. The pipe loop system was dominated by advection and
thus showed little variance in the values of BF. Analysis of the series of pressurized tank systems
showed significant regions of turbulent mixing and recirculation corresponding to a system that
was much less efficient than the pipe loop system. The open surface (plastic) tank systems
exhibited the most uneven flow paths and lowest efficiencies (BF) seen in this study. These
systems exhibited significant degrees of short-circuiting and recirculation largely due to their
inlet and outlet configurations.
Link
Page 43
The minimum flow velocity in pipe loop contactors is calculated using the Reynolds Number. In
order to receive full credit, the pipe loop contactor must operate in the turbulent flow regime. To
achieve this, the Reynolds Number of the flow must be greater than 4000.
where:
= System flow rate in cubic feet per second (CFS)
Re = 4000
= kinematic viscosity of water (
if the water is at 70F)
= 3.14
D = pipe diameter in feet
Example:
Pipe Size: 6 inches (0.5 feet)
Water temperature: 70F
Table 9: Physical System Specifications (note this pipe loop contactor had no bends within
the system)
Pipe Length
Pipe Diameter
System Volume
L/D Ratio
Page 44
75 feet
6 inches
110.2 gallons
150
Re
Flow Rate (cfs)
Flow Rate (GPM)
Velocity (ft/s)
TDT (seconds)
Laminar
1500
.00589
2.64
0.03
2500
0.7
Turbulent
15000
.0589
26.44
0.3
250
0.9
BF*
* Note: The BF for the turbulent pipe loop contactor is 0.9 and not 1.0 that would be awarded for
this system. This is because a BF = 1.0 is not an exact fit for the system in question. As more
piping is added to this system, the BF would asymptotically approach 1.0.
Figure 23: Pipe Loop Contactor RTD Curves from Laminar/Turbulent Study
Page 45
Re
Flow Rate (cfs)
Flow Rate (GPM)
Velocity (ft/s)
TDT (seconds)
Turbulent
15000
0.0589
26.44
0.3
250
Table 12: Physical System Specifications (note this pipe loop contactor had no bends within
the system).
Pipe Length
Pipe Diameter
System Volume
TDT
BF
Page 46
L/D Ratio
40
150
20
75 feet
6 inches
6 inches
29.4 gallons
110.2 gallons
67 seconds
250 seconds
0.7
0.9
Figure 24: Pipe Loop Contactor RTD Curves from L/D Study
Page 47
Page 48
Figure 65: Physical Tank Geometry (Left) and Modeled Tank Geometry (Right)
CFD models were developed and simulated using ANSYS WorkBench and ANSYS Fluent in
order to estimate the average velocity profile and hydraulic disinfection efficiency of the basic
system displayed in
Figure 6. This process was repeated for two additional cases, which are outlined in Figure 7. All
geometry was created and meshed using software attached to ANSYS WorkBench and then
imported into ANSYS Fluent for analysis. Flow and scalar transport were modeled as outlined in
Appendix F, specifically Section 2.4.2 on page 12 in Flow Dynamics and Scalar Transport in
Drinking Water Contact Tanks, which is not described here for the sake of brevity. Also similar
to the study conducted by Barnett (see Appendix F), a rigid lid model was applied. Each scenario
was run at a specified flow rate of 20 gallons per minute and the flow depth was estimated using
physical measurements and data summarized in Appendix F, specifically Section 3.3.3 on page
30 in Flow Dynamics and Scalar Transport in Drinking Water Contact Tanks found. This was
found to be around 5.23 feet (62.7 in) for all cases.
Within each model a passive scalar was continuously injected at the inlet of the tank and a flux
averaged monitor was placed at the outlet. Data resulting from this setup yields a representative
RTD curve that can be used to estimate the baffling factor (for more information on RTD curves
and baffling factors please see Section 2.2 on page 6 in Flow Dynamics and Scalar Transport in
Drinking Water Contact Tanks found in Appendix F). These numerical RTD curves were
compared to results from physical tracer studies in order to validate the outlined methodology.
Page 49
Physical tracer studies were conducted in a controlled lab setting using a full sized prototype.
Flow was brought to steady state and then tracer solution was continually injected into the flow
using a constant displacement pump. Tracer was fully integrated before entering the tank with
the use of a static mixing tube. Effluent was monitored with the use of a fabricated flow through
device and sampling times were based off of observed results from the CFD models. Two
separate tracers were used in this study: sodium chloride and lithium chloride. Sodium chloride
was used for the majority of the physical tracer studies and was measured indirectly with the use
of a YSI EC300A conductivity meter. Lithium chloride was used sparingly due to financial
Page 50
constraints. For lithium chloride tests samples were taken at predetermined time intervals via a
diversion valve on the effluent and then analyzed in the Soil, Water, and Plant testing laboratory
at CSU using inductively coupled plasma-atomic emission spectroscopy. For more information
on applied testing procedures see the document entitled Improving Drinking Contact Tank
Hydraulics using Random Packing Material listed in Appendix H.
Both CFD models representing the base case from
Figure 6 and Additional Case II from Figure 7 were validated with the use of physical tracer
studies. Physical tracer studies were conducted for the base case using sodium chloride and
physical tracer studies were conducted for Additional Case II using sodium chloride and lithium
chloride. The physical construction of Additional Case II can be seen in
Figure 8. The normalized plots of the study results can be seen in Figure 9 and Figure 10. CFD
results and physical tracer study results agree exceptionally well, validating the proposed
computational models.
Page 51
Page 52
Once the applied CFD methodology was validated the derived velocity fields could be used to
design potential applications of random packing material. Contours of velocity for the three
modeled cases can be seen below in Figure 30 - 32. Displayed units are
where is
the point velocity magnitude and
is the average velocity magnitude of the plotted tank.
All contours are plotted on a plane that intersects the center of the inlet. Areas that are shaded red
represent velocity magnitudes that are five times the average or greater.
It can be seen from Figure 30 that the jet created by the use of a sharp inlet induces significant
amounts of flow separation and unused tank volume. For the base case velocities within the jet
become as high as 40 times the average tank velocity, causing it to quickly cut through the tank
volume, hit the back wall, and rapidly spread towards the outlet. Using a sharp inlet causes flow
to undercut the water column, leaving the majority of the tank unused. The RTD curve in Figure
9 displays this trend, as they break through almost instantaneously. Such significant short
circuiting has been observed in other open tanks with sharp inlets (see e.g. Appendix F,
specifically Section 5.3 on page 71 in Flow Dynamics and Scalar Transport in Drinking Water
Contact Tanks and Appendix C, specifically Section 3.2.3 on pages 58 in Evaluation of Flow and
Scalar Transport Characteristics of Small Public Drinking Water Disinfection Systems using
Computational Fluid Dynamics found in).
The introduction of baffles into the tank mitigates some of the severe short circuiting by forcing
the flow through a number of channels, but the presence of a high velocity jet is enough to
induce short circuiting and significant flow separation in several channels. For the one baffle
case (see Figure 31) the velocity distribution within the second channel is still significantly
skewed and non-uniform. The resulting baffling factor is around twice that of the base case, but
it is still less than 0.10 and the resulting curve breaks through quickly. For the two baffle case
(see Figure 32) the velocity field approaches uniformity only in the last channel and the first two
channels still contain significant amounts of short circuiting. Flow separation at the first baffle
turn is also quite significant. The use of a sharp inlet significantly detracts from the use of
internal baffling. Tanks using weir inlets (see Appendix G, specifically Section 3.6 on page 52 in
Towards Improved Understanding and Optimization of the Internal Hydraulics of Chlorine
Contact Tanks) show significantly larger gains in efficiency from the use of long baffles than the
present case simply because they have more favorable inlet conditions. Results suggest that
modifying the inlet and the baffle turns could potentially increase the performance of the tested
systems.
Page 53
Plan View
Figure 31: CFD Velocity Field: Alternate Case I: One Baffle (
above)
Page 54
) (Views same as
Figure 32: CFD Velocity Field: Alternate Case II: Two Baffles (
above)
) (Views same as
Based off of the velocity fields displayed in Figures 30-32, a parametric study was conducted in
order to determine the befits of placing a box of random packing material at the tank inlet. Box
dimensions were varied in order to determine an optimal design for dispersing the resulting jet
caused by a sharp inlet. Investigated dimensions include box length
and box height
.
For all tested configurations the box width
was set at the width of the tank (four feet). A
three dimensional schematic of a typical inlet box can be seen in Figure and photographs of
several tested boxes can be seen in
Figure 34. Each investigated box will be described in this appendix by using the notation
where both dimensions are in feet. A list of tested
Page 55
Page 56
Figure 34: Inlet Box Prototype Examples (From Left to Right: 1ft X 2ft, 1ft X 4ft, 2ft X 2ft)
Each box was tested at flow rates of 10, 20, and 40 GPM using the methods described in section
E.3 of this appendix. Sodium chloride was used almost exclusively as a tracing agent, but lithium
chloride was used for the 1ft X 2ft box case in order to validate the applied methodology. Results
of this validation case can be seen in Figure 11.
Page 57
Table 13 highlights several key concepts. First, possible gains in hydraulic disinfection
efficiency with only the use of an inlet box are limited. BF values do not exceed 0.40 regardless
of parameter manipulation. Also, the BF appears to decrease with the box length and with flow
rate. Since
is located parallel to the predominant trajectory of the jet, it is expected that
increasing
will allow for more effective dissipation. This trend can be observed, but
asymptotic behavior in measured BF values suggests that only a finite amount of dissipation can
occur and that beyond a certain point increasing
yields little additional gain (there is a
threshold). As the flow rate decreases the measured baffling factor also decreases. The greatest
decrease occurs between the flow rates of 10 GPM and 20 GPM. As the flow rate drops the
amount of energy contained in the turbulent jet decreases. For flow rates below 20 GPM it
appears that the jet does not contain enough energy to be effectively dissipated (it will not be as
dispersed). Benefits diminish as the flow approaches the laminar flow regime. The height
appears to have little effect on the BF of the system.
Figure 126 - Figure 137 provide examples of parameter effects.
Page 58
at 20 GPM (Right)
Work presented in section E.5 showed that the application of random packing material as an inlet
diffuser could increase the baffling factor of the studied tank by around 620%, but this still
leaves the BF at a lower value of 0.36. With this in mind an additional set of physical tracer
studies was conducted in order to investigate the application of random packing material within a
baffled system. Studies considered the placement of random packing material at areas of high
velocity and flow separation (at the inlet and at baffle turns). The previously validated system
containing two baffles was used as a modified case study. Based off of results from section E.5, a
1ft X 4ft box was placed at the inlet of the baffled tank. After this situation was fully analyzed,
boxes were constructed and placed over the entire opening of each baffle turn. Images of each
physical set up can be seen in Figure 148. Results indicated that the attempted modifications
were significantly beneficial, but only for flow rates above a certain minimum threshold.
Figure 148: Inlet Box and Turn Boxes within Baffled System
Page 59
Results from the study can be seen in Figure 39. Applying an inlet box and turn boxes within the
baffled system at 20 GPM yields a baffling factor of 0.55, which is 10 times greater than the
initial baffling factor. However, the observed gains are significantly sensitive to flow rate. Figure
40 and 41 show the effect of flow rate on the baffling factor of the full setup. As the flow rate
decreases the flow at the inlet approaches the laminar-turbulent transition regime and the energy
of the flow decays. Beneath 20 GPM the observed benefits of the baffle turns diminish and the
BF drops to around 0.40. It is possible that at a flow rate of 10 GPM the amount of energy in the
sharp inlet jet is not sufficient to productively overcome the resistance of the packing material.
At this point the packing material may even induce additional separation and reduce effective
flow area. The use of industrial packing material in open surface concrete tanks can significantly
increase hydraulic mixing efficiency, but it must be applied under the correct conditions.
Page 60
Page 61
Page 62
Page 63
NOTE: This article is available for purchase from American Water Works Association. To
purchase the article please visit http://dx.doi.org/10.5942/jawwa.2014.106.0005. The abstract to
the article is shown below.
This study investigated the use of industrial packing material for increasing the hydraulic
efficiency of small-scale drinking water chlorine contact tanks. Packing material used in this
study was spherical with porosities between 0.9 and 0.95, and a density less than water. It should
be noted that the void ratios has been dropped to down to 0.8 in Section 7 of the guidance
document to allow for less porous packing material than those tested in this study to be used. To
be conservative (in response to this relaxation in void ratios), the baffling factors shown in Table
8 in Section 7 have been assigned conservatively. Sixty-seven tracer studies were conducted on
laboratory scale chlorine contact tank systems exploring three sizes of packing material, two tank
sizes, and two flow rates. Sodium chloride solution was injected as a continuous tracer at the
inlet and monitored in the tank outflow through electrical conductivity. Several studies were
validated with the use and direct measurement of a lithium ion tracer. Hydraulic efficiency was
measured by determining the baffling factor as outlined by the US Environmental Protection
Agency (USEPA). Results suggest the utilization of packing material in small drinking water
contact tanks can significantly increase the baffling factor, thus improving the disinfection
efficiency obtained from the existing tanks.
where:
Veff = effective tank volume (%)
BF = system baffling factor
Vtank = volume of tank without packing material (gallons)
Vpacking = volume of packing material to be used (gallons)
porosity = the porosity (or void fraction) specified by the manufacture
Page 64
The Department and CSU have jointly developed a tracer test protocol to be used as an example
for water systems to conduct tracer testing. The link below contains the example test protocol.
Tracer testing may be necessary to expand plant capacity, justify a current baffling factor, or to
better understand your process at a water plant. The Department relies heavily on the guidance
published by the USEPA for performing tracer studies. The document is titled, Guidance
Manual for Compliance with the Filtration and Disinfection Requirements for Public Water
Systems using Surface Water sources (1991). Tracer testing is covered in Appendix C. The
exception is that the Department will allow the water system to test at two flow rates: >90% of
peak flowrate and near 50% peak flowrate.
Link to Tracer Protocol
As part of the overall scope of work, CSU conducted two full scale tracer studies at public water
systems. The results of these tracer studies are presented in the documents below.
Link to CSU Tracer Study, Jamestown
Link to CSU Tracer Study, TVW Water System
Page 65